INDEX

Explanations

actions related to movement or direction

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ont

-0.15

kt

-0.15

asm

-0.14

Occ

-0.14

Ã¥l

-0.14

abez

-0.14

vas

-0.13

ly

-0.13

aling

-0.13

POSITIVE LOGITS

chwitz

0.17

 orth

0.15

ãģ

0.15

 Predictor

0.15

ORTH

0.14

ynn

0.13

phia

0.13

osten

0.13

adc

0.13

inh

0.13

Activations Density 0.004%

actions related to movement or direction

No Comments

No Known Activations